University of California, San Francisco Logo

University of California, San Francisco | About UCSF | Search UCSF | UCSF Medical Center

Home > Global Health Literature Digest > Outcomes of Botswana's National Antiretroviral Treatment Program
Five-year outcomes of initial patients treated in Botswana's National Antiretroviral Treatment Program
Global Health Sciences Literature Digest
Published May 11, 2009
Journal Article

Bussmann H, Wester CW, Ndwapi N, et al. Five-year outcomes of initial patients treated in Botswana's National Antiretroviral Treatment Program. AIDS 2008;22:2303-11.

In Context

Botswana is the first country in sub-Saharan Africa to have provided antiretroviral therapy (ART) and, as such, is able to report on long-term outcomes. To date, over 90,000 people have received treatment through the national ART program. Although ART has been shown to have excellent immunologic and virologic response, short-term mortality rates among patients treated in the region are higher than those found in developed countries.(1, 2, 3, 4)


To measure clinical outcomes and death among people receiving ART


The Infectious Disease Care Clinic (IDCC) at Princess Marina Hospital (PMH); Botswana's largest hospital, which is located in its capital, Gaborone

Study Design

Prospective cohort study


All ART-naïve patients registered at the IDCC between January 21, 2002 and August 7, 2002


Mortality, change in CD4 counts and HIV viral load, change to second-line ART, adherence


The Botswana national treatment guidelines for adults recommend two nucleoside reverse transcriptase inhibitors (NRTIs), zidovudine/lamivudine, and one non-nucleoside reverse transcriptase inhibitor (NNRTI). Typically, the NNRTI is efavirenz or, for women of reproductive age, nevirapine. The regimen consisted of 200 mg of Combivir (lamzid, once available) twice daily. Nevirapine began with a two-week lead-in period of 200 mg once a day followed by a maintenance regimen of 200 mg twice a day. Efavirenz was dosed at 600 mg each day.

Patients experiencing virologic failure on first-line therapy were switched to a protease inhibitor-based regimen (lopinavir/ritonavir) coupled with two NRTIs (didanosine and stavudine or tenofovir or abacavir and lamivudine). When virologic failure occurred with second-line therapy or when there were complicated first-line therapy failures, genotypic resistance testing was undertaken.

At the initial visit, all patients underwent a comprehensive physical examination and additional diagnostic tests for comorbid conditions (e.g., chest radiographs for suspected tuberculosis) and baseline laboratory tests that consisted of chemistry, hematology, CD4 cell counts, plasma HIV-1 RNA, hepatitis B surface antigen, and syphilis serology. Opportunistic illnesses were managed according to national guidelines. Prophylaxis against opportunistic illnesses consisted of cotrimoxazole for patients with CD4 counts less than 200 cells/mm3 and isoniazid/pyridoxine for all patients.

Clinically eligible patients were started on ART within two weeks of registration. People treated with nevirapine had liver function tests done prior to dose increases at two weeks. CD4 and viral load tests were conducted quarterly through November 2006 and semi-annually thereafter. Chemistry and hematology test results were obtained after four weeks and after three months. Beginning in 2005, patients had chemistry and hematology, lipase, non-fasting and fasting glucose, lipid, and lactase testing as needed.

Confirmatory tests were drawn for patients with HIV-1 RNA levels of more than 400 copies/mL and these patients also received adherence counseling and education. Specially trained physicians provided care in the first year and trained nurses thereafter. Patients were provided with counselors trained to assist with adherence and to identify toxicity to ART. Patients obtained monthly refills at the pharmacy where pill counts were performed and where patients were provided with additional counseling and education. Clinical data were collected initially using paper records followed by comprehensive electronic records that included clinical and laboratory data.

Data from all patients were included in the analysis. Observations were censored at April 1, 2007 or at the time of last clinic visit or transfer. Time to virologic failure and changes in regimens were censored at the date of last clinic visit, and the time to laboratory endpoints was censored at the date of the last collection of laboratory specimens. The Kaplan-Meier product limit method was used to estimate the time-to-events distribution. Log-rank tests were used to compare outcomes after ART across groups. Likelihood ratio tests from Cox proportional hazards models were used to compare outcomes after ART initiation for continuous variables and multiple outcome predictors. For analyses involving HIV RNA or CD4 counts, the data were grouped into intervals around selected time points centered on each time period and the measure that was closest to the time point was selected; patients contributed one observation to each interval or no observation. Predictors of death or loss to follow-up were measured using the log-rank test.


There were 871 patients enrolled at PMH IDCC between January 21, 2002 and August 7, 2002. Of these, 633 ART-naïve adults initiated therapy. The median age was 34.8 years, 60% were women and 38% were WHO clinical stage 4 and 43% were stage 3. The median duration of follow-up was 41.9 months and the total duration of follow-up was 1822.2 patient-years. The median duration of follow-up for the 290 patients alive and known to be receiving ART as of April 1, 2007 was 57.2 months. The Kaplan Meier estimates of loss to follow-up for all 633 patients at one, three, and five years were 8.9%, 15.4%, and 21.9%, respectively.

The increases in median CD4 cell count at one, three, and five years were 169, 303, and 337 cells/µL, respectively. For patients whose baseline CD4 counts were <50 cells/µL, the median increases at one, three, and five years were 200, 315, and 367 cells/µL, respectively, while for patients with CD4 counts above 50, the median increases were 162, 298, and 326 cells/µL, respectively.

Undetectable viral loads were observed for 91.3%, 90.1%, and 98.3% of patients at one, three, and five years, respectively. Virologic failure occurred among 47 patients, and the Kaplan Meier estimates of failure were 17.6%, 22.1%, and 30.1% at the three time points.

There were 120 deaths during follow-up. Kaplan-Meier survival estimates at six months and one, three, and five years were 87.8%, and 82.7%, 79.3%, and 79.0%, respectively. Survival estimates for patients with baseline CD4 counts of <50 cells/µL were lower: 74.8%, 71.1%, and 70.5% at these three time points. One-year mortality was 2.3-fold greater among patients who started treatment at a baseline CD4 count of <50 cells/µL compared with patients whose started therapy at counts between 51 and 200 cells/µL. Predictors of mortality included baseline CD4 counts of <50 cells/µL, WHO clinical stage 3 to 4, and hemoglobin levels below 8 g/dL. Data on cause of death were determined using clinical records and verbal autopsy information. Use of these data sources provided information on 71 (59.2%) patients, for whom the most common causes of death were advanced HIV disease with wasting syndrome (34.2%) and pulmonary tuberculosis (6.7%) as the lead opportunistic illness.

Serious medical problems that resulted in a change of ART at six months were anemia (6.9%), cutaneous hypersensitivity reaction (1.3%), and hepatotoxicity (1.1%). Adherence was measured by the frequency of medication refill visits at the pharmacy beginning in July 2002--the point at which computerized pharmacy records were available for tracking refill visits. The mean overall monthly pharmacy refill rate was 92.5%, and 73.7% of all patients missed zero to one refill visit.


The findings from this study indicate the clinical benefit of ART coupled with adequate adherence and minimal need for regimen changes.

Quality Rating

The quality of the study is very good. The cohort is large and representative of people in the region who initiate therapy. The five-year follow-up is longer than follow-up in other studies of clinical outcomes with ART in this region. Clinical and adherence measures were appropriate. The loss to follow-up and transfer-out rates are high and, as such, the outcome from these patients is not known. It may be that patients who were lost to follow-up died, in which case the survival benefit would have been over-estimated.(5) For patients who transferred care to another treatment site, it could be argued that their outcomes would be similar to those who remained at PMH.

Programmatic Implications

The findings from this study demonstrate that ART has clear clinical and survival benefits. Given that the disease severity in this cohort was high, one could expect better outcomes in patients who initiate therapy earlier. Additional comorbid conditions in this region, such as poor nutrition and tuberculosis, are likely to contribute to the deaths observed. The evidence for successful treatment of HIV-infected people in Botswana is substantial. Additional efforts to diagnose HIV with prompt referral for clinical care and provision of ART to patients both with less-advanced and advanced disease must be prioritized


  1. Stringer JS, Zulu I, Levy J, et al. Rapid scale-up of antiretroviral therapy at primary care sites in Zambia: feasibility and early outcomes. JAMA 2006;296:782-93.
  2. Wester CW, Kim S, Bussmann H, et al. Initial response to highly active antiretroviral therapy in HIV-1C-infected adults in a public sector treatment program in Botswana. J Acquir Immune Defic Syndr 2005;40:336-43.
  3. Coetzee D, Hildebrand K, Boulle A, et al. Outcomes after two years of providing antiretroviral treatment in Khayelitsha, South Africa. AIDS 2004;18:887-95.
  4. Braitstein P, Brinkhof MW, Dabis F, Schechter M, Boulle A. Mortality of HIV-1-infected patients in the first year of antiretroviral therapy: comparison between low-income and high income countries. Lancet 2006;367:817-24.
  5. Wester CW, Okezie O, Thomas AM, et al. Higher-than-expected rates of lactic acidosis among highly active antiretroviral therapy-treated women in Botswana: preliminary results from a large randomized clinical trial. J Acquir Immune Defic Syndr 2007;46:318-22.